Assitance needed: Please check my maths.

I have learned that Watts = volts x amps.
So one Cree q5 uses a fv of 3.7V. So Driven at 950mah = 3.515 Watts.
So 7 Cree q5 is 3.15 watts x 7 = 24.605 Watts?

This is Driven by the blue shark convertor board which is approx 90% efficient so the loss is 10% which = 2.46 watts?

So Cree's + Blue shark should use approximately = 28 watts?

Still with me.....

The 7 Cree should pull from the shark a total Forward voltage of 3.7 x 7 which = 25.9 total forward voltage?

Hang on in there....

I want to use a 6 cell 4000mah 22V Lipo battery which has a max hot of the charger voltage of 4.2V (6x4.2) = 25.2.

1. Is this just about going to be ok as the blue shark is a boost only driver and i dont want to kill it, expensive to UK.

2. So how do i work out how much current will be being drawn from the battery to check it is below the blue sharks 4A max input level?

3. Assuming all the above is correct? I think my battery then has 88WH? So my theretical run time would be 88 mwh / 28 watts = 3 hours?

Pheww.. Please feel absolully free to correct me in any way you like.

I just needed to know that i can use 7 cree with a blue shark and get 2 - 3 hour run times with a battery that fits in a cylinfer with internal diammeter of 62mm, if the above all works then i think i have cracked it yippeee....

Thanks.

your assumption is a little off

gav6280 said:

I want to use a 6 cell 4000mah 22V Lipo battery which has a max hot of the charger voltage of 4.2V (6x4.2) = 25.2

Click to expand...

@ 1A draw use about 3.7V per cell it's much more accurate for LiPo (even if it's 1/4C)

1. OK for blue shark.
2. about 1.3A ~~~ current = watt / volt = 28 / (6* 3.7)
3. 88 Wh ... what's with the MegaWattHour?

oh yes , sorry well spotted was wh not mwh, that would be a dam nig battery eh?

So my sharks input would go from 1.12A to 1.55 as the voltage falls away to 18V?

Giving an average [SIZE=+1]1.387 Amps per Hour?

Battery at 4000mah / 1387mah = [/SIZE][SIZE=+1]2.88 hours.:twothumbs
[/SIZE]

2.88 hours would be the longest burn time you could theoretically get (assuming the above math is correct). But since you are pulling a relatively high current from the battery, and the battery may not always actually hold a full charge due to age and self discharge etc you may want to consider it as a 2 hour capacity. But then if you have it dimmable then run time will be increased.

At that high a battery voltage, the Shark wouldn't be able to dim the light. You didn't say if that's a feature of interested to you.

Nope, just on or off needed.

Cheers.

Im still a little concerned that if my total forward voltage was to dip slightly due to the shark board not putting out it's full capacity,then i would end up in direct drive and fry yet another! blue shark. And they aint cheap to get in the UK.

Theretical lowest forward voltage of led string = 25.2
Theretical Max voltage of a 6 cell battery straight of charge = 25.44

So could i not just add a 1 Ohm 3 watt resistor in series into my string to raise the forward voltage by 1 volt, just to be on the safe side?

Or would i be better putting a resistor between the battery and the shark?

I think i have it sorted out, im going to use a silicone diode to give an additional 0.7 drop in the forward voltage.

http://www.gavinwilkinson.com/circuit%20jpeg.jpg

What do you think?

I was thinking about your design last night - a well thought out, well balanced supply / draw design. Sometimes these are the hardest to manage.

In the case of a single LiIon cell and a low Vf LED, the cell voltage passes through the LED Vf - making it seem like you need a buck / boost setup. You are getting close to that and carefully avoiding it.

Adding the diode is the right approach to increase Vf in the LED string.

There is one more option which I tend to like, and others tend to not be impressed with - use a small resistor - 1 - 2 ohms.

When you have a large drive voltage like your setup and relatively little voltage drop to be deal with (V Bat - Vf) then a resistor is pretty handy. The power loss might be as little as 30% of the arrangement you have, with nearly flat output.

The minimum Vbat is going to actually be more like 6 x 3.5 = 21 volts, and even then, you will be at 80 - 90 % light output. Dropping the current 10 % will not result in 10 % light drop due to heating / cooling effects.

7 x power LEDs in a light - should be a nice bright package. Just an FYI - it is not so easy to get rid of more than 10 watts of heat without real thought. Keep the heat away from the LiPos.

Yeah thought of that, i hope it's ok though because the led's are mounted on a solid chunk of 11mm thick aluminum 63mm in diammeter which is press fitted into a aluminum pipe with 3mmm thick walls.

Oh and did i forget to mention.....It's going underwater and will have the whole ocean to help cool it.

I just have to hope and pray enough of the heat makes the transfer to the casing though....

At the moment the back of the heat sink has the shark board mounted on it to. but it's only got an 5mm air gap before the start of the battery.

Any ideas for a insulating material that i can cut a disc out of and epoxy onto the end of the battery, it can be a maximum of 5mm thick. (What's that stuff they put on space shuttles - Anyone got any)

I think i am going to with a silicone diode to drop the voltgae by 0.7 as they dont use any current like a resistor, unless someone can give mem a good reason why not?

or then again wouldnt that resistor idea also bring down the current and thus bring down the heat a smidge.....

Now you got me thinking....ohhhh....So confused....

Think i may have just answered that one.

http://cgi.ebay.co.uk/MX-aluminium-exhaust-heat-shield-sheet-sticker-pad_W0QQitemZ250341422283QQcmdZViewItemQQptZUK_Motorcycle_Parts?hash=item250341422283&_trksid=p3286.c0.m14&_trkparms=66%3A2|65%3A2|39%3A1|240%3A1318

I meant skip the shark driver completely and just use a 1 or 2 ohm resistor instead.

Underwater, you will actually be better off with individual colors mixing than the white LEDs.

For penetration in seawater, best is 488nm if the real goal is penetration, not color rendering.

The lack of sufficient red in the LEDs will show up strongly in underwater photography if that is the goal. The spectral output of a white LED is quite different than the sensors of most cameras - the digital ones are nearly all RGB with a few adding an extra cyan sensor. White LEDs are nearly all blue / yellow with some extra red thrown in.

Adding some color LEDs needs to be done with planning through, the non-blue ones will typically have a lower Vf than white.

Maybe that is more than you wanted to know.

Good stuff, but your right. That would be like starting again and it's already on it's 3rd re design, and im yet to actually get a single one finished which is now the priority....